primary tumor size-dependent inhibition of angiogenesis at a
TRANSCRIPT
(CANCER RESEARCH 58, 5866-5869, December 15. 1998]
Primary Tumor Size-dependent Inhibition of Angiogenesis at a Secondary Site: AnIntravital Microscopic Study in Mice1
Axel Sckell,2 Nina Safabakhsh,3 Marc Dellian,4 and Rakesh K. Jain5
Department <>fRadiation Oncology. Hauoduutttl General Hospital. Hiiminl Medical School. Boston. Massachusetts 02I14
ABSTRACT
Some primary tumors are capable of suppressing the growth of theirmétastasesby presumably generating antiangiogenic factors such as an-
giostatin. We hypothesized that the amount of inhibitor(s) released by atumor increases with tumor growth. We tested this hypothesis by evaluating the relationship between the size of a primary tumor and its abilityto inhibit angiogenesis at a secondary site. Furthermore, we characterizedthe effects of the primary tumor on physiological properties of newlyformed vessels at the secondary site. Angiogenesis and physiological properties were measured using intravital microscopy of angiogenic vessels inthe gels containing basic fibroblast growth factor placed into cranialwindows of immunodeficient mice bearing human prostatic carcinoma(PC-3) in their flank. The PC-3 tumor inhibited angiogenesis in the gels,
and surgical resection of tumor reversed this inhibition. The inhibition ofangiogenesis 20 days after gel implantation (range, 0-83%) correlatedpositively (r = 0.625; /' < 0.008) with the tumor size on the day of gelimplantation (range, 19-980 mm"1).The primary tumor also suppressed
leukocyte-adhesion in angiogenic vessels, thus helping them evade the
immune recognition. These results provide an additional rationale forcombining antiangiogenic treatment with local therapies.
INTRODUCTION
Although complete removal of primary tumors normally is the mosteffective therapy to reduce the risk of metastatic disease, in certaincases removal of primary tumors is accompanied by a sudden accelerated growth of preexisting métastasesat secondary sites ( 1). Growthof métastasescan also be inhibited by tumor mass (2). Furthermore,the growth of a second tumor graft can be hindered by a primarytumor (3). Based on these observations, Folkman and co-workers (4)
proposed the following hypothesis: a primary tumor, although capableof stimulating its own vascular bed, can inhibit angiogenesis in thevascular bed of a metastasis or other secondary tumor (4). O'Reilly et
al. (4. 5) have proven this hypothesis by discovering angiostatin (4)and endostatin (5). These substances specifically inhibit endothelialcell proliferation in vitro. They also inhibit angiogenesis in vivo notonly in tumors but also in angiogenic gels implanted in the cornea ofmice (4-6).
We propose here that the amount of angiogenesis inhibitors generated by a primary tumor should increase with tumor size. Therefore,the extent of angiogenesis inhibition at a secondary site should increase with tumor size. A prerequisite to test this hypothesis is theavailability of an assay that permits in vivo quantification of bothangiogenesis inhibition and tumor size. We have recently developed
Received 7/2/98; accepted 10/15/98.The costs of publication of Ihis article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate Ihis fact.
1Supported by National Cancer Institute Outstanding Investigator Grant R35-
CA56591 (to R. K. J.). M. D. and A. S. are recipients of Fcodor Lynen-Fellowships fromthe Alexander von Humholdl-Foundation. D-53173 Bonn. Germany.
•Present address: Department of Orthopedic Surgery. Inselspital. University of Berne.CH-3010 Berne. Switzerland.
1Present address: Department of Internal Medicine. Temple University Hospital.
Temple University, Philadelphia, PA 19140.4 Present address: Department of Otorhinolaryngology, Head and Neck Surgery. Klini
kum Grosshadem. University of Munich, D-81366 Munich. Germany.' To whom requests for reprints should be addressed, at Department of Radiation
Oncology, COX-7. Massachusetts General Hospital. Boston. MA 02114.
such an assay (7, 8). In this assay, a gel containing bFGF6 is placed
in the cranial window implanted in a mouse. Using intravital microscopy, we can measure angiogenesis and physiological properties ofthe newly formed vessels in the gel. Thus, we can obtain a betterunderstanding of physiological and pathophysiological events duringangiogenesis inhibition at a secondary site. We designed this study totest the following hypotheses: (a) the size of the primary tumorcorrelates with its antiangiogenic potency at a secondary site; (b)removal of the primary tumor reverses its antiangiogenic effects; and(c) the primary tumor adversely affects the hemodynamics, leukocyte-
endothelium interaction, and microvascular permeability of newlyformed vessels in the gel.
MATERIALS AND METHODS
Animal Model, Anesthesia, Surgical Techniques, and Tumor Implantation. The experiments were performed in SCID mice (age. 6-10 weeks;
mean body weight, 25.0 ±3.2 g) bred and housed in our defined flora animalcolony. A mixture of ketamine hydrochloride (Ketalar; Parke-Davis. MorrisPlains, NJ; 100 mg/kg body weight) and xylazine (Xyla-Ject: Phoenix Phar
maceutical, St. Joseph, MO; 10 mg/kg body weight) was used as anesthesia. Inone set of male SCID mice (PC-3 mice), a piece (=1 mm in diameter) of ahuman hormone-independent prostatic carcinoma (PC-3; Ref. 9) was im
planted s.c. into the right flank. In controls, no tumors were implanted. Tofacilitate surgical resection of tumors in tow at a later time, tissue-isolated
tumors were grown in four female SCID mice. These tumors are connected tothe host-vasculature with a single ovarian artery and a single ovarian vein. Tothis end, the left ovary was positioned s.c. and injected with a 0.03-ml slurryof minced PC-3 tissue in HBSS (H-9269: Sigma Chemical Co., St. Louis, MO)
into its fat pad (10). A stretched paraffin film (Parafilm; American CanCompany. Greenwich. CT), wrapped around the ovary, prevented inftltrativetumor growth. In all animals, the preparation of the cranial window wasperformed as described earlier (11).
Angiogenesis Assay. The preparation of the angiogenic gel and its implantation into the cranial window have been described in detail elsewhere (7). Inbrief, 20 ¡uof a gel containing 60 ng of human recombinant bFGF (LifeTechnologies, Inc.. Gaithersburg. MD) was sandwiched between two squarepieces of nylon mesh (=2.5 X 2.5 mm; Tetko, Briarcliff Manor, NY) and
placed on the pia mater inside the cranial window. bFGF was dissolved in 2.4fil of 0.1% BSA (BSA; Sigma). 6.5 mg of aluminum sucrose octasulfate(sucralfate; courtesy of BM Research a/s, Va;rl0se, Denmark), and 17.6 /j.1oftype I collagen (Vitrogen 100: Collagen Corp.. Palo Alto. CA) neutralizedpreviously to pH 7.4 by the addition of 1 part of sodium bicarbonate ( 11.76 g/1)and 1 part of lOx minimal essential medium (Life Technologies, Inc.) to 8parts Vitrogen 100. The mesh contained 25 to 35 squares with 300-/xm
interval.Quantification of Angiogenesis. For quantification of angiogenesis, the
anesthetized animal was placed under a stereomicroscope (SMZ-1: Nikon,Tokyo, Japan) with epi-illumination and 12.5-fold magnification (7). The
angiogenic response (R) in the gel. expressed in percentage, was calculated asthe ratio between the number of squares of the top nylon mesh containing atleast one newly formed vessel (JVVCS)and the total number of squares (JV10I):R = 100 X NvcJNtM {%]. The inhibition of angiogenesis on day 20 after gelimplantation (/) was calculated as follows: / = 100 —R [%].
Intravital Microscopy and Permeability Measurement. In vivo microscopy was performed using a microscope with epi-illumination (Axioplan;
'' The abbreviations used are: bFGF. basic fibroblast growth factor; SCID, severe
combined immunodeficient: Rho-BSA. tetramethylrhodamine-labeled BSA.
5866
Research. on February 12, 2018. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from
INHIBITION OF ANGIOGENESIS BY A PRIMARY TUMOR
Zeiss, Oberkochen, Germany) equipped with a X20 long-working-distance
objective (LD Achroplan. X 20/0.40 korr; Zeiss) and a fluorescence filter set(Omega Optical, Brattleboro, VT). For off-line analysis, regions of interestwere recorded on video tapes using a S-VHS videocassette recorder (SVO-9500 MD; Sony, Tokyo, Japan) and a regular charge-coupled device (CCD;AVC-D7; Sony, Tokyo, Japan) or intensified CCD video camera (C2400-68:Hamamatsu Photonics K.K., Hamamatsu, Japan). FITC-labeled dextran (Sig
ma; M, 2,000,000; 0.1 ml of a 5% solution in 0.9% NaCl as a plasma marker)and rhodamine 6G (Molecular Probes, Eugene, OR; 0.04 ml of a 0.05%solution in 0.9% NaCl as a fluorescence dye to label WBC in vivo) wereinjected i.v. to visualize the microcirculation and to measure leukocyte-endo-thelium interaction, respectively. The effective permeability (P) to tetrameth-ylrhodamine-labeled BSA (Rho-BSA; Molecular Probes; 0.8 mg/ml PBS) of
newly formed vessels in the angiogenic gels was measured as describedpreviously (11, 12). The fluorescence intensity of one selected area per animalwas measured with a photomultiplier (9203B; Thorn EMI, Rockaway, NJ)before and 1, 3, 5, 7, 9, 11, 14, 17, and 20 min after i.v. injection of a 0.12 mlbolus of Rho-BSA.
Off-Line Analysis of Microhemodynamics and Leukocyte-Endothelium
Interaction. Inconsistency in microhemodynamic parameters can stem frommechanical or morphological properties of vessels. To avoid this problem, wechose only confluent, venous-like 100-¿unvessel segments with a straight
course and similar diameters (15.5 ±2.3 /im) from five to eight regionsselected randomly per animal and time of investigation. Each of these regionswas recorded for 30 s on video tape for the evaluation of leukocyte-endothe-
lium interaction and for 60 s for hemodynamic measurements. Internal diameters (D) in ¡urnand RBC velocity (VRBC)in /j.m/s were measured using adigital video image-shearing monitor (model 908; IPM, San Diego, CA) and a
four slit photometric analyzer (model 208C; IPM), respectively, both connected to a personal computer (IBM PS/2,40SX; Computerland, Boston, MA).The number of sticking leukocytes (Ns) is expressed as the density of cells(WBCS) firmly adhering to the inner surface of the investigated 100-fim vesselsegment: WBCS = IO6Nsl(ir X D x 100) [cells/mm2]. Rolling leukocytes are
defined as a population of cells temporarily interacting with the vessel wall andthus having a velocity at least 50% smaller than VRBC.Rolling count (WBCR)is given as a percentage: WBCR = 100 X /VRWwBc [*]•where NR is thenumber of rolling leukocytes and A'WBC 's tne number of all nonstickingleukocytes. Leukocyte flux (FWBC) was normalized by the cross-section area
of the observed microvessel and the observation time (/):FWBC = 10" X ATWBC/[(D/2)2X ir X /] [cells/mm2/s]. Apparent wall shear rate(y) was derived from Poiseuille's law for Newtonian fluid: y = (vRBC/D) X 8
[1/s]. Blood flow (Q) was calculated as follows: Q = VRBCX D2 X ir/4000
[pl/s].Analysis of Tumor Volume. The size of all tumors implanted was meas
ured using a caliper and corrected for skin thickness (0.5 mm in SCID mice).Tumor volume (VTU) was obtained using the formula: Vj^ = TT!6 X DmaJl X (Dmin)2 [mm1], where Draax is the maximal tumor diameter and
Dmin is the corresponding perpendicular diameter.Autopsy and Histology. After sacrificing the animals at the end of exper
iments, all mice with tissue-isolated tumors as well as some randomly selectedmice with PC-3 implanted s.c. were chosen for autopsy to scan for métastases.
In these animals, lung, liver, spleen, kidney, heart, and spine were cut in slicesfor macroscopic investigation. The tissue samples were fixed in 10% formaldehyde and embedded in paraffin. For microscopic investigations, tissuesections were cut and stained with H&E.
Animal Groups. Matched animal groups were used for all experiments.The sequence of measurements between controls and PC-3 mice was selectedrandomly. In PC-3 mice, the angiogenic gel was implanted when the tumor had
reached the desired size. However, before gel implantation in animals, aminimum time of 7 days was allowed to elapse after cranial window preparation to recover from the surgery. Animals with surgery-induced brain bleed
ing, neurological symptoms, or any sign of inflammation were excluded fromthe study. To find out whether the effect of PC-3 on angiogenesis response inthe gels is reversible, PC-3 was implanted in four animals as tissue-isolated
preparation and was removed 2 weeks after gel implantation.Statistics. Results are presented as means ±SD unless specified otherwise.
Data were analyzed statistically using one-way repeated measures ANOVA.As a multiple comparison test, Bonferroni's Method was applied if a normality
test was passed and Friedman repeated measures ANOVA as well as Student-
Newmans-Keuls Method when normality test failed, respectively. The Pearson
Product Moment Coefficient was calculated to analyze correlations. For alltests, Ps <5% were considered significant.
RESULTS
Quantitative Angiogenic Response and Growth of PC-3. In 17animals bearing PC-3, the inhibition of angiogenesis on day 20 after
gel implantation correlated positively with the tumor volume on theday of gel implantation (Fig. 1/4). Of these 17 animals, angiogenicresponse in gel assays of 7 PC-3 mice with tumor volumes >205 mm1
on the day of gel implantation was compared with 22 controls (Fig.Iß).Beginning on day 10 after bFGF gel implantation, there was asignificant inhibition of angiogenesis in gels in PC-3 mice compared
with the control mice. On day 20 after gel implantation, the differencein the angiogenic response between the two experimental groupsbecame 61%. In contrast to controls where angiogenesis significantlyincreased over time, in PC-3 mice from day 14 onwards the angio
genic response seemed to be suppressed at a constant level. Fig. 1Cshows the growth of PC-3 of the above-mentioned seven PC-3 micewith tumors >205 mm3 on the day of gel implantation.
Effect of Tumor Removal on Angiogenic Response. Surgicalremoval of tissue-isolated PC-3 in four animals 14 days after gelimplantation proved that tumor-induced inhibition of angiogenic re
sponse to bFGF was completely reversible (Fig. ID). Within 14 daysafter tumor resection, angiogenesis increased from 29.4 ±9.1% to98.8 ±2.5%.
Autopsy and Histology. In all mice autopsied, macroscopic aswell as microscopic investigation of lung, kidney, spleen, liver, heart,and spine did not reveal any sign of tumor métastasesas an alternativesource for the production of growth-inhibitory substances. All organs
investigated showed normal morphology according to the age of themice.
= B§zoE«>—
4ËiA100
-75
-50
-«25
:OH^•«t*f10
300•0•¡
lì jo600
900VOLUME [mml
E 3000 -
W 2000-
3 1000 -o> o
F D-100 H
S375H50H
7 10 14 17 20
TIME [d]
Tumor Removal
0 7 10 141720
TTME[d]
7 14 21 2528
TIME [d]
Fig. 1. Analysis of angiogenesis and tumor growth. A, correlation of the inhibition ofangiogenesis on day 20 after gel implantation in PC-3 mice (n - 17) and their tumorvolume at the day of gel implantation (r = 0.625, P < 0.008: •.single measurements).B, angiogenic response in angiogenesis gel assays at different times after gel implantationof controls (D; n = 22) and PC-3 mice <•:n - 7) bearing tumors >205 mm' on the day
of gel implantation (*. P < 0.05 versus controls; #. P < 0.05 versus previous measurement). C. tumor growth of the PC-3 mice shown in B (§,P < 0.05 versus all measurements 7 days or earlier). D. angiogenic response in angiogenesis gel assays of mice(n = 4) bearing tissue-isolated PC-3 before and after tumor removal (#, P < 0.05 versusprevious measurement; •.single measurements). Bars, SD.
5867
Research. on February 12, 2018. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from
INHIBITION OF ANOIOOENESIS BY A PRIMARY TUMOR
7 14 21 7 14 21
TIME after Gel Implantation [d]
Fig. 2. Microhemodynamics and leukocyte-endothelium interaction in angiogenicvessels of controls (D; n = 7) and PC-3 mice (•;n = 7). A. RBC velocity. B, shear rate.C blood flow. D, leukocyte flux. £.rolling count. F, sticking leukocytes. »,P < 0.05versus controls. Bars, SD.
Microhemodynamics and Leukocyte-Endothelium Interaction.Analysis of RBC velocity (Fig. 2A), shear rate (Fig. 2ß),and bloodflow (Fig. 2Q revealed no differences between PC-3 animals and
controls over 21 days after gel implantation. At comparable leukocyteflux (Fig. 2D) in both groups of animals on days 7, 14, and 21 aftergel implantation, rolling count (Fig. 2E) in PC-3 mice on average was
decreased to 28, 27, and 22% of the corresponding control values. Thenumber of sticking leukocytes in PC-3 mice (Fig. 2F) showed a
tendency of being reduced, reaching the level of significance only onday 14 after gel implantation. The age of the newly formed vessels inthe angiogenesis gel assay had no effect on leukocyte-endothelium
interaction.Microvascular Permeability. No differences were found in mi-
crovascular permeability between PC-3 mice and controls at any
given time (Fig. 3). In contrast, in both animal groups the very youngvessels on day 7 after gel implantation revealed on average 3.4- and2.0-fold higher permeability to Rho-BSA than older vessels 7 and 14
days later, respectively.
DISCUSSION
The present data show that PC-3 can nonspecifically inhibit angio
genesis at a secondary site. Furthermore, in support of our hypothesis,the size of the primary tumor correlated positively with its antiangio-
genic effects at a secondary site. The angiogenesis inhibition wascompletely reversed after surgical resection of the primary tumor. InPC-3 mice during inhibition of angiogenesis, no changes in microhe-modynamic parameters and permeability of newly formed microves-sels were detected, whereas leukocyte-endothelium interaction was
suppressed.Animal Model. The in vivo angiogenesis gel assay and mecha
nisms responsible for the bFGF-induced angiogenesis have been
discussed in detail elsewhere (7, 13). The present animal model offersa new approach with a high spatial and temporal resolution to performextensive qualitative and quantitative analysis of antiangiogenic effects induced by a primary tumor at a secondary site. In a similar
model where sucralfate pellets containing bFGF were implanted intocorneal micropockets of mice, microscopical investigations were performed only on days 5-7 after pellet implantation (6). The present
model allows monitoring of the time course of angiogenesis over atime period of up to 4 weeks.
Angiogenic Response. Angiostatin is most likely the substanceresponsible for the PC-3-induced antiangiogenic effects observed in
the present study (14). The positive correlation between tumor volumeon the day of gel implantation and inhibition of angiogenesis on day20 after gel implantation (Fig. \A) gives strong evidence for a dosedependence of the antiangiogenic effects; larger tumors producehigher amounts of antiangiogenic substances. Animals bearing PC-3smaller than 205 mm3 showed the same angiogenic response as
controls (data not shown). In a few PC-3 mice bearing large tumors,
regression of some already established microvessels took place (datanot shown). This is in accordance with observations made by O'Reilly
et al. (5), where systemic administration of endostatin induced regression of primary tumors to dormant microscopic lesions. It was notpossible to follow PC-3 mice with tumors >205 mm3 at the day of gel
implantation for >3 weeks. Because of the fast and invasive growthof PC-3, the animals had to be sacrificed after this time. Thus, wecould not ascertain as to how long primary tumor-induced suppression
of angiogenesis would last. In the present study, surgical removal oftissue-isolated PC-3 abrogated the antiangiogenic effects seen intumor-bearing animals similar to the observations in some patients
and animals (1,2, 9). Note that the amount of bFGF in gels to induceangiogenesis in our mouse model is relatively much higher than theamount of angiogenic factors produced by early-stage métastasesinhumans. Therefore, the relatively huge size of PC-3 tumors is neces
sary to suppress the angiogenic response in the gels sufficiently and,hence, should not be extrapolated on a weight-for-weight basis to the
clinical situation in patients.Microhemodynamics and Leukocyte-Endothelium Interaction.
Leukocytes (i.e., mast cells, macrophages, neutrophils, and lymphocytes) are normally associated with angiogenic processes duringwound healing, inflammation, and tumor growth (15-17). They have
the potential of secreting a wide variety of substances for stimulatingangiogenesis and cell proliferation such as proteases, growth factors,and many other monokines (18, 19). Recruitment of leukocytes fromcirculation into extravascular space is dependent on a multistep cascade of events involving rolling, sticking, and emigration (17, 20, 21).The reduced leukocyte-endothelium interaction in the gels in PC-3
mice at a comparable leukocyte flux in both experimental groupscould in part explain why the angiogenic response was suppressed inprimary tumor-bearing animals. However, the exact mechanism of
action of antiangiogenic substances such as angiostatin and endostatinstill remains unclear. RBC velocity, shear rate, and blood flow did notdiffer between controls and PC-3 mice or over the time within each
experimental group. Therefore, it is likely that the factors produced bythe primary tumor (e.g., bFGF and transforming growth factor ß)induced the reduced leukocyte-endothelium interaction in PC-3 mice
7 14 21TIME after Gel Implantation [d]
Fig. 3. Microvascular permeability (P) of angiogenic vessels in controls (D; n = 8) andPC-3 mice (•;n = 8). #, P < 0.05 versus corresponding group on day 7 after gel
implantation. Bars, SD.
5868
Research. on February 12, 2018. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from
INHIBITION OF ANGIOGENESIS BY A PRIMARY TUMOR
via reduced expression and/or loss of a number of adhesion moleculeson the surface of newly formed vessels in the angiogenesis gels. Atsimilar microhemodynamic parameters and leukocyte flux, in controlsthe number of leukocytes interacting with the endothelium of newlyformed microvessels in the angiogenesis gel assay was comparablewith data of previous microcirculatory studies in tumors and normaltissues (22). Thus, primary tumor may facilitate evasion of immunerecognition at the secondary site by suppressing leukocyte-endothelial
interaction.Microvascular Permeability. In general, microvascular perme
ability observed in the angiogenesis gels was much higher than innormal tissues and similar or slightly higher than in neoplastic tissues(11, 12, 23, 24). No differences were detected between PC-3 mice and
controls. Permeability measurements on days 14 and 21 were comparable with previous measurements on days 12 and 25 in the samesystem (7). Additionally, we found significantly higher values for thisparameter on day 7 after gel implantation versus days 14 and 21.Although there was no direct statistical correlation between micro-
vascular permeability and angiogenic response, it seems that vessels ata very early stage of development are more leaky than more maturevessels. This facilitates the extravasation of proteins, which are important for the formation of extracellular matrix for further cellmigration (25).
Conclusion. In summary, our model is suitable for investigatingthe effects of growth-inhibitory factors produced by a primary tumor
on angiogenesis at a secondary site. The observed positive correlationbetween the size of a primary tumor and inhibition of angiogenesis ata secondary site as well as the reversibility of this effect after tumorresection supports the hypothesis that local eradication or regressionof certain primary tumors may enhance angiogenesis and thus growthof preexisting métastaseselsewhere. Therefore, a cancer patientshould benefit from combined local and long-term antiangiogenicfollow-up therapies.
ACKNOWLEDGMENTS
We thank Sylvie Roberge for providing the tissue-isolated tumors, Ed Rose
and his staff for animal care, and Carol Lyons for help with manuscriptpreparation.
REFERENCES
1. Sugarbaker. E. V., Thomthwaite, J.. and Ketcham. A. S. Inhibitory effect of a primarytumor on metastasis. In: S. B. Day. W. P. L. Myers, P. Stansly, S. Carrettini, andM. G. Lewis (eds.). Progress in Cancer Research and Therapy, pp. 227-240. NewYork: Raven Press, 1977.
2. Prehn, R. T. The inhibition of tumor growth by tumor mass. Cancer Res., 51: 2-4,1991.
3. Gorelik, E. Resistance of tumor-bearing mice to a second tumor challenge. CancerRes., 43: 138-145, 1983.
4. O'Reilly, M., Holmgren, L., Shing, Y., Chen, C. Rosenthal, R. A., Moses, M.. Lane.
W. S., Cao, Y., Sage, E. H., and Folkman, J. Angiostatin: a novel angiogenesis
inhibitor that mediates the suppression of métastasesby a Lewis lung carcinoma. Cell.79: 315-328, 1994.
5. O'Reilly, M., Boehm, T., Shing, Y.. Fukai, N., Vasios, G., Lane, W. S.. Flynn, E.,
Birkhead. J. R.. Olsen. B. R., and Folkman. J. Endostatin: an endogenous inhibitor ofangiogenesis and tumor growth. Cell, 88: 277-285, 1997.
6. Chen, C., Parangi, S., Tolentino, M. J.. and Folkman, J. A strategy to discovercirculating angiogenesis inhibitors generated by human tumors. Cancer Res.. 55:4230-4233, 1995.
7. Dellian, M., Witwer. B. P., Salehi. H. A., Yuan, F.. and Jain. R. K. Quantitäten andphysiological characterization of angiogenic vessels in mice: effect of basic fibroblastgrowth factor/vascular permeability factor and host microenvironment. Am. J.Pathol., 149: 59-71. 1996.
8. Jain, R. K., Schlenger. K.. Hockel, M.. and Yuan. F. Quantitative angiogenesis assays:progress and problems. Nat. Med.. .). 1203-1208, 1997.
9. Kaighn. M. E., Shankar Narayan. K.. Ohnuki. Y.. Lechner, J. F., and Jones, L. W.Establishment and characterization of a human prostatic carcinoma cell line (PC-3).Invest. Urol., 17: 16-23, 1979.
10. Kristjansen, P. E. G., Roberge. S., Lee, I., and Jain, R. K. Tissue-isolated humantumor xenografts in athymic nude mice. Microvasc. Res., 48: 389-402. 1994.
11. Yuan, F.. Salehi. H. A., Boucher, Y., Vasthare, U. S.. Turna, R. F.. and Jain. R. K.Vascular permeability and microcirculation of gliomas and mammary carcinomastransplanted in rat and mouse cranial windows. Cancer Res., 54: 4564-4568. 1994.
12. Yuan, F.. Leunig. M.. Berk. D. A., and Jain. R. K. Microvascular permeability ofalbumin, vascular surface area, and vascular volume measured in human adenocar-cinoma LS174T using dorsal chamber in SCID mice. Microvasc. Res., 45: 269-289,
1993.13. Senger. D. R. Molecular framework for angiogencsis: a complex web of interactions
between extravasated plasma proteins and endothclial cell proteins induced by angiogenic cytokines. Am. J. Pathol.. 149: 1-7, 1996.
14. Galley, S.. Twardowski, P., Stack, M. S., Patrick. M.. Boggio. L.. Cundiff, D. L.,Schnapper. H. W.. Madison, L., Volpert. O.. Bouck. N.. Enghild. J.. Kwaan. H. C.,and Soff. G. A. Human prostate carcinoma cells express enzymatic activity thatconverts human plasminogen to the angiogenesis inhibitor, angiostatin. Cancer Res..56: 4887-4890, 1996.
15. Leibovich, S. J., and Ross, R. The role of macrophages in wound repair. A study withhydrocortisone and antimacrophage serum. Am. J. Pathol., 78: 71-100. 1975.
16. Leek, R. D., Lewis, C. E., Whitehouse. R.. Greenall. M.. Clarke. J.. and Harris, A. L.Association of macrophage infiltration with angiogenesis and prognosis in invasivebreast carcinoma. Cancer Res., 56: 4625-4629, 1996.
17. Jain, R. K., Koenig, G. C., Dellian. M., Fukumura, D., Munn, L. L., and Melder, R. J.Leukocyte-endothelia! adhesion and angiogenesis in tumors. Cancer Metastasis Rev.,/5: 195-204, 1996.
18. Sunderkötter, C., Steinbrink. K., Goebeler, M., Bhardwaj, R., and Sorg, C. Macrophages and angiogenesis. J. Leukocyte Biol., 55: 410-422. 1994.
19. Cullino, P. Prostaglandins and gangliosides of tumor microenvironment: their role inangiogenesis. Acta Oncol., 34: 439-441, 1995.
20. Butcher, E. C. Leukocyte-endothelial cell recognition: three (or more) steps tospecificity and diversity. Cell, 67: 1033-1036, 1991.
21. Springer, T. A. Traffic signals for lymphocyte recirculation and leukocyte emigration:the multi step paradigm. Cell, 76: 301-314, 1994.
22. Fukumura, D., Salehi, H. A., Witwer, B.. Turna, R. F.. Melder. R J., and Jain, R. K.Tumor necrosis factor a-induced leukocyte-adhesion in normal and tumor vessels:effect of tumor type, transplantation site, and host strain. Cancer Res., 55: 4824-
4829, 1995.23. Gerlowski, L. E., and Jain. R. K. Microvascular permeability of normal and neoplastic
tissues. Microvasc. Res., 31: 288-305. 1986.
24. Yuan, F., Chen, Y., Dellian. M.. Safabakhsh, N.. Ferrara, N., and Jain, R. K.Time-dependent vascular regression and permeability changes in established humantumor xenografts induced by an anti-vascular endothelial growth factor/vascularpermeability factor antibody. Proc. Nati. Acad. Sci. USA, 93: 14765-14770, 1996.
25. Dvorak, H. F., Brown. L. F., Detmar. M.. and Dvorak, A. M. Vascular permeabilityfactor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am. J. Pathol., 146: 1029-1039, 1995.
5869
Research. on February 12, 2018. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from
1998;58:5866-5869. Cancer Res Axel Sckell, Nina Safabakhsh, Marc Dellian, et al. Secondary Site: An Intravital Microscopic Study in MicePrimary Tumor Size-dependent Inhibition of Angiogenesis at a
Updated version
http://cancerres.aacrjournals.org/content/58/24/5866
Access the most recent version of this article at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
.http://cancerres.aacrjournals.org/content/58/24/5866To request permission to re-use all or part of this article, use this link
Research. on February 12, 2018. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from